We are currently carrying out an epidemiologic survey of 35-64 year old Mexican residents of several low-income colonias in Mexico City in which we expect to enroll a total of 296 subjects by Feb 28, 1993. Preliminary results indicate that, compared to age-matched, low-income Mexican Americans in San Antonio, Mexicans in Mexico City may have "carbohydrate-induced lipemia" (i.e., high carbohydrate diet, hypertriglyceridemia and low HDL). Moreover, Mexican diabetics are more hyperglycemic than Mexican American diabetics. These findings raise questions about the relative rates of cardiovascular disease and microvascular complications of diabetes in Mexicans vs. Mexican Americans. To address these questions we propose to carry out a 3.5 year prospective followup of our Mexico City cohort to determine the incidence of type II diabetes. We anticipate examining 1952 of the original 2296 subjects. All participants will receive a medical history interview and a health habits interview, the latter covering diet, physical activity, cigarette smoking, and alcohol use. Participants will also undergo a clinic examination which will include a glucose tolerance test with insulins, lipids and lipoproteins including measurement of "small, dense LDL", blood pressure,, anthropometrics, and an EKG. Baseline and follow-up EKGs will be Minnesota-coded to assess both prevalence and incidence of EKG abnormalities. Special examinations to further assess cardiovascular disease and microvascular complications of diabetes will include: ankle-arm blood pressure ratios to assess peripheral vascular disease to be performed on all subjects by Doppler ultrasound; ultrasound examination to assess carotid disease in 400 randomly selected subjects; and fundus photography to assess diabetic retinopathy in the anticipated 298 diabetics and 100 non-diabetic controls. Fundus photographs will be read by the University of Wisconsin Reading Center. Diabetic nephropathy will be assessed by clinical proteinuria and microalbuminuria in the same subset of patients. All prevalence and incidence rates will be compared to the corresponding rates for Mexican Americans and non-Hispanic whites in San Antonio in whom all of the above described measurements are currently being made using comparable methodology.